With the rise of Internet and other related technologies, system requirements for servicing ever increasing network traffic have dramatically changed. Ever more powerful server systems, which are often configured as an array of servers, are often provided to service requests originating from external sources such as the World Wide Web, for example. As local Intranet systems have become more sophisticated thereby requiring servicing of larger network loads and related applications, internal system demands have grown accordingly as well. Unfortunately, as demands for more complicated and higher performance systems have increased, conventional tools and/or methods for measuring, predicting, and planning whether a given server is capable of servicing a desired network load have not been adequate.
Presently, many performance measurement systems rely on a connection model in order to determine whether a server may have the requisite capacity to service a desired network load. According to the connection model, a plurality of client systems may be configured to generate a large number of concurrent connections. However, this type of model may not adequately stress a given server and/or application and is thus inadequate for a number of reasons. One such problem is the connection model does not provide a controllable amount of stress for the server. Since each client operates independently, a suitable mechanism, for example, would be needed to coordinate the various load requests. Unfortunately, such mechanisms are presently unavailable. Another problem relates to providing a sustained and/or predictable rate of requests. Generally, since clients are generating load requests open loop (e.g., without monitoring the number of requests), peaks and valleys of requests are likely to occur. Therefore, clients may likely wait while outstanding connections remain open wherein a given server may not be stressed to its fullest capacity. Thus, performance measurements based on an inconsistent load are difficult to determine.
Still another problem associated with conventional systems relates to adjusting desired loads. Currently, in order to increase the number of load requests, an additional client system may have to be installed to provide additional connections. Consequently, costs are increased and resulting granular adjustment of combined server load difficult to attain. Still, conventional systems that provide loosely coupled/controlled and/or unregulated output stream of requests do not enable accurate predictions to be made as to whether an application and/or system may operate at a desired capacity or stress level. In view of the above problems, there is an unsolved need for a system and/or methodology to generate a consistent network load, provide a more realistic load model, and thereby improve server performance testing to achieve improved capacity planning.